Cloud Scraper

This bot attempts to discover publicly accessible media by generating random short codes for a variety of hosting services and checking which links are valid. Any discovered content is posted to a Discord channel.

Configuration

Copy config.example.json to config.json and fill in your Discord bot token and the channel ID you want to post found images to. Optionally provide Matrix credentials to send messages to Matrix rooms. The bot looks for this file in the same directory as bot.py, so ensure it is placed there even if you launch the script from a different working directory.

The bot has a built-in list of supported services so no additional configuration is required.

Supported services

Domain	Code length
prnt.sc	6
tinyurl.com	6
is.gd	6
bit.ly	7
rb.gy	6
app.goto.com/meeting	9 digits
reddit.com	6 (posts parsed for media link)

Reddit posts are treated like redirects to the linked image or video and the media is uploaded directly when possible.

{
  "token": "YOUR_DISCORD_BOT_TOKEN",
  "channel_id": 123456789012345678,
  "scrape_workers": 4,
  "matrix_homeserver": "https://matrix.org",
  "matrix_user": "@bot:matrix.org",
  "matrix_password": "YOUR_MATRIX_PASSWORD",
  "matrix_rooms": ["!roomid:matrix.org"]
}

matrix_rooms is optional. If omitted, the bot will post to every room it has joined and will automatically join new rooms when invited. Omit the Discord or Matrix credentials entirely to disable posting to that platform.

Running

Install the requirements and run bot.py:

pip install -r requirements.txt
playwright install
python bot.py

The bot will log attempts and post any discovered images to the configured Discord channel and/or Matrix rooms.

Character frequency statistics are saved to char_stats.json and used to bias code generation toward more common letters for each domain. Only successful codes contribute to this file. The file is loaded when the bot starts and written back regularly alongside the domain weights so learning persists between runs. Additional heuristics about letter case and digit placement are recorded in pattern_stats.json using only successful codes.

Each domain also maintains a simple weight that influences how often it is selected for testing. Domains start at 1.0 and increase by 0.1 whenever a link is valid. Invalid links decrease the weight by 0.01, but a domain's weight will never drop below 1.0. The current weights are stored in domain_stats.json so the bot can learn over time which services are more reliable.

If either of these files is missing or contains invalid JSON, the bot will reset them to default empty structures on startup.

Concurrency

The scraper can run multiple workers in parallel to speed up scanning. The number of workers can be configured in config.json using the scrape_workers field (defaults to 4). You can also override this at runtime with the SCRAPE_WORKERS environment variable:

SCRAPE_WORKERS=4 python bot.py

Each worker launches its own browser instance, so increasing this value will use more system resources.